基于跟踪微分器的波门跟踪算法的研究

摘要

In this paper,an improved dual-window tracking algorithm based on tracking differentiator filter is proposed to meet the practical requirement in real-time and precision of the target tracking system. The possible position of moving target center in the next frame image is predicted by TD filter. A search region for target is generated around the target center position. Simulation results illustrate that the proposed algorithm has superior features,such as its independences to model information, higher prediction precision. It also can meet the practical requirement in real time processing.%针对目标自动跟踪系统对目标跟踪算法实时性和精确性要求高的特点,提出了一种基于跟踪微分滤波器的运动目标跟踪方法.该算法利用跟踪微分器预测目标中心在下一帧体现在图像中可能出现的位置,以该位置为波门目标检测的中心,减少了目标的搜索范围,缩短了算法的运算时间.仿真结果表明,该方法具有不依赖于目标运动模型,实时性好,精确性高等优点.